p.1 #1 · Sharpen D800 Image or De-moire D800E Image
I've read through many of the forums and it is clear there is confusion regarding the pros/cons of the D800 and D800E.
Can someone share their thoughts about the benefits of sharpening (unsharp mask) an image produced by a D800 to create an image similar to a D800E? Is it possible to create a nearly identical result without risk of moire?
p.1 #2 · Sharpen D800 Image or De-moire D800E Image
The D800E image would have more information so it will always be sharper, all other things being equal. The D800 cannot match the D800E image. You can't add what isn't there.....
p.1 #3 · Sharpen D800 Image or De-moire D800E Image
You can remove AA filter and there are places that can do that for you. A lot of Pros did and do such conversion. Normally you pay for it $500 so if you get D800E you will actually save $200. Also your warranty won't be voided.... (but lets say you do this after the warranty expires)
p.1 #6 · Sharpen D800 Image or De-moire D800E Image
Hi Omume
I agree with the prior posts. The D800E will be sharper and whatever you could do post processing with the D800 you could match with post processing with a D800E.
That said, I personally would go with the D800. The D800E is more prone to anti-aliasing artifacts including Moiré. I do not want to have to worry about artifacts whether repetitive Moiré strips on a large scale or small adjacent pixel aberrations that could defeat the purpose of the extra sharpness.
Here is some info that might help put the value of the extra sharpness into perspective. from examples that Nikon has provided of apples to apples sharpness comparisons between the D800 and D800E, it appears that the extra sharpness is on the order of about 1/2 pixel (some quick measurements in Photoshop). To put this in perspective, if you blew up an image to 40x60 inches and viewed it from 10 inches away (pretty extreme) two side by side 1/2 pixels would create a spot that is only 0.2mm in size. This size is widely recognized as the threshold of perception by the average human eye.
So unless I am going to need to cover extreme cropping and/or extreme enlargements with expectations of very close viewing of those large prints, I don't think that practically, I will see the advantage in most of my work. So the advantage for those extreme conditions for me do not outweigh the likelihood of extra post processing for for anti-aliasing created Moiré which can occur independent of print size.
p.1 #7 · Sharpen D800 Image or De-moire D800E Image
Actually .03 microns is the benchmark and the max print size is d800 36MP, pixels per image = 7360 x 4912 so max print at 300dpi is 24.53 x 16.37 inches.
The sensor has a pixel pitch of 4.88µm quite small drastically reducing DoF but increasing resolution. Unless you shoot textiles I would personally go with a D800E, however for the inexpensive price of the body you may want to consider both bodies and slap a prime on either.
That being said I am convinced both will be a great travel camera and your snapshots will be tack sharp providing you have the appropriate glass. I look forward to the non biased review by someone whom doesn't link there site to others for kickbacks.
Also please be advise that you will not want to remove the (3) filters in front of the sensor Moire' is not the only issue you will have to deal with,there will be light rays bouncing off the micro lenses causing specular highlights. There is no software made that can debunk your file after you remove the 3 filters before the sensor.
Feb 14, 2012 at 01:31 PM
Guest
Guest
p.1 #8 · Sharpen D800 Image or De-moire D800E Image
2613pch wrote:
Actually .03 microns is the benchmark and the max print size is d800 36MP, pixels per image = 7360 x 4912 so max print at 300dpi is 24.53 x 16.37 inches.
The sensor has a pixel pitch of 4.88µm quite small drastically reducing DoF but increasing resolution. Unless you shoot textiles I would personally go with a D800E, however for the inexpensive price of the body you may want to consider both bodies and slap a prime on either.
That being said I am convinced both will be a great travel camera and your snapshots will be tack sharp providing you have the appropriate glass. I look forward to the non biased review by someone whom doesn't link there site to others for kickbacks.
Also please be advise that you will not want to remove the (3) filters in front of the sensor Moire' is not the only issue you will have to deal with,there will be light rays bouncing off the micro lenses causing specular highlights. There is no software made that can debunk your file after you remove the 3 filters before the sensor.
taken from a thread i started at LL
"DoF is proportional to format, not pixel pitch. Your images will not look softer compared to a lower resolution Nikon 35mm sensor--if your image is sharp, why is dividing the image into smaller pixels going to change that?
This idea of linking DoF and pixel pitch comes from pixel peeping. 100% monitor views are in no way a real world condition for viewing images and gives a exaggerated view of DoF and sharpness"
p.1 #9 · Sharpen D800 Image or De-moire D800E Image
John Wheeler wrote:
That said, I personally would go with the D800. The D800E is more prone to anti-aliasing artifacts including Moiré. I do not want to have to worry about artifacts whether repetitive Moiré strips on a large scale or small adjacent pixel aberrations that could defeat the purpose of the extra sharpness.
There is more to it than just resolution. Micro-contrast is also better on the D800E. As image viewers, it is micro-contrast that more strongly gives us a sense of sharpness. The micro-contrast in a D800E cannot be gained in post processing an D800 version of the image.
So unless I am going to need to cover extreme cropping and/or extreme enlargements with expectations of very close viewing of those large prints, I don't think that practically, I will see the advantage in most of my work. So the advantage for those extreme conditions for me do not outweigh the likelihood of extra post processing for for anti-aliasing created Moiré which can occur independent of print size.
For some subjects and shooting styles, a D800E is a bad idea. It is difficult to fix moiré in post, and the results end up being worse than in an AA filter had prevented it. However, there is no guarantee that moiré will never appear on the D800. Moiré can be found in images from Nikon DSLRs that have AA filters.
It looks like the D800 will outsell the D800E by 2:1 or greater.
My take is that Nikon had a very good reason to make two cameras. Neither can exactly match the other's capabilities. The difference in image quality between the D800E and D800 (not counting moiré) will be subtle. I've noticed that some of the high-end photographers who have blogged about the D800/E seem to recommend the D800 for most people, and then mention that they have ordered the D800E.
If you do studio work with a tethered camera and can review images as you shoot, you might lean towards the D800E. Also, if you shoot natural landscapes from a tripod, the D800E is a good choice because moiré causing subject matter rarely shows up.
Cityscapes, and fabrics are a good source for moiré problems. So a D800 could be a better choice there.
p.1 #10 · Sharpen D800 Image or De-moire D800E Image
Hi 2613pch
I used to live near Newport Beach. Have relatives still in the area.
Actually .03 microns is the benchmark and the max print size is d800 36MP, pixels per image = 7360 x 4912 so max print at 300dpi is 24.53 x 16.37 inches.
I believe you are referring to the Circle of Confusion (CoC) as it is quoted for a full frame 35mm sensor which is .03mm (I bet yet meant that instead of micron if I understand what you were saying). That .03mm number is based on what the average person can see when the image is enlarged to an 8x10 inch print and viewing from a distance of 10 inches. That is a 7.5X factor (if you use diagonals of the sensor to 8x10 diagonal) and that yields .225mm (I rounded to .2mm). That equates to 5 line pair per mm which is the average persons acuity when viewing an image from 10 inches. All of the DOF calculators including DOF Master use the same approach. They compare the diagonal of an 8x10 to the diagonal of the sensor and scale .225mm down by the same factor. That is why 35mm is .03mm, Canon APS-H sensors are .023mm Nikon DX are .02mm and Canon APS-C are .019. And for reference, the CoC for an 8x10 camera for the same approach is .225mm (i.e. 1.0X scale factor).
This level of extra information is rarely included with all the DOF calculators. The DOF of what you can perceive is sharp is also a function of the enlargement of the print and your viewing distance from viewing the print. The assumptions about .03 mm also assume that you are not cropping the image either.
I agree about the max print size based on 300 dpi. My point was that IMHO to see the improved sharpness of the D800E which I believe is smaller than the dimension of a pixel, my calculations indicate that you would have to enlarge to 40x60 inches and view it at 10 inches to see the difference. All based on the same .2mm. The apples to apples examples that Nikon showed on their website show the magnitude of improvement. You can see it, I just question whether you see that smaller prints. I think it will require actual use/tests and prints to see the real differences. That will come with time.
The seminal reference for all this DOF stuff is "Photogrphic Optics" 1950 by Allen R Greenleaf. If gives a good explanation of the details. This is the reference to which all the DOF calculators eventually refer. Copies are available through used online book stores. I got a copy a while back. Very enlightening so encourage anyone that wants to know details about optics including DOF, diffraction etc, I really recommend it. Then again, we may just want to take pictures instead
To me, that's a noticeable difference - but the big question is what is the difference once the image is printed.
ACNYPhoto wrote:
I'd really like to see more D800/e comparison shots to really see if the lack of AA filter is a good/bad thing...
I agree, I would like to see a lot more examples and testing to see the extra sharpness of the D800E. Getting a good apples to apples comparison is not easy to do given a lot of other contributing factors in lack of sharpness (e.g. was same exact physical lens used, how far off center from lens, focus point, exact same exposure parameters)
Steve the image you linked to is intriguing.
It looks obviously sharper. Couple things I noticed
- The size by side comparison before being magnified are 100% views. The magnified view of the side by sides are at 300%
- Unless there was some wind that blew in a branch in the lower right corner, the two shots were at different vantage points (made me wonder what else was different).
Another item that also has to be taken into account is DOF in the tests. For landscape scenes, that you want to focus out to infinity yet maximize DOF, my understanding is that you focus at the hyperfocal distance and you will then be "in focus" from 1/2 the hyperfocal length to infinity. Where you have to be careful with these tests is that the focus point needs to be identical. Using the standard DOF calculators, with a FF sensor the Circle of Confusion (CoC) used is .03mm (unless you change that in the calculator). This means that at both the 1/2 hyperfocal distance and at infinity you will have a blur that is .03mm on the sensor. Note that is 30 microns and the pixel size of the D800/D800E is 4.8 microns. The blur circle using these standard techniques is basically 6 pixels wide for the D800/D800E. My point being is that if comparing the D800 to the D800E, if you don't have the exact same focus point for the two comparison shots you have another uncertainty with which to deal.
This also means that if you want that extra sharpness of the D800E to be noticeable, the DOF range for that level of sharpness should be calculated with a CoC set to about .005 IMHO.
Fortunately, by the time I save up enough money to afford a D800E, all of this testing will have been done and I won't have to depend on guessing or just sample images with undocumented comparison criteria. It will be very "clear" by then.
p.1 #15 · Sharpen D800 Image or De-moire D800E Image
afm901 wrote:
The D800E image would have more information so it will always be sharper, all other things being equal. The D800 cannot match the D800E image. You can't add what isn't there.....
Right. You can't add what isn't there. The extra sharpness on the D800E is not there; it is only apparent. The aliasing that an Anti-Aliasing filter prevents is the generation of false data as spatial frequencies fold back.
The D800, like all modern DSLRs, will have a weak AA filter and will show some aliasing but not as badly as the D800E.
Sharpening (a D800) in post is also creating data that doesn't exist in the scene (dips and spikes around edges), but the data is based on real data and has a perceptual basis and is not semi-random data generated by aliasing.
In most cases, however, the lenses (even the sharpest Nikkors) are going to challenge the D800 / D800E sensors but the sensors will out-resolve the lenses about 2:1 at 50 % MTF (estimated). This means they meet the Nyquist frequency sampling criteria.
Thus in almost all cases, the Nikkor will act as the AA filter.
p.1 #16 · Sharpen D800 Image or De-moire D800E Image
One important question to ask when comparing D800 and D800E images is whether they are derived from in-camera conversion of the Raws to JPEGs or whether the conversion is down outside and what outside software converted it (Nikon or Adobe). Nikon may be applying slightly different processing to in-camera JPEGs between the two cameras (first and second cases) and their outside software as a third case. Adobe as a fourth case will not have exactly the same algorithms.
p.1 #18 · Sharpen D800 Image or De-moire D800E Image
Monito wrote:
Thus in almost all cases, the Nikkor will act as the AA filter.
turnert wrote:
Good point. Which is why I will likely add the 'E' to the kit
That's odd. The conclusion that I would derive from the quoted analysis is that the extra money for the E would be a waste and the D800 would be the camera to add.
p.1 #19 · Sharpen D800 Image or De-moire D800E Image
Monito wrote:
Right. You can't add what isn't there. The extra sharpness on the D800E is not there; it is only apparent. The aliasing that an Anti-Aliasing filter prevents is the generation of false data as spatial frequencies fold back.
The D800, like all modern DSLRs, will have a weak AA filter and will show some aliasing but not as badly as the D800E.
Sharpening (a D800) in post is also creating data that doesn't exist in the scene (dips and spikes around edges), but the data is based on real data and has a perceptual basis and is not semi-random data generated by aliasing.
In most cases, however, the lenses (even the sharpest Nikkors) are going to challenge the D800 / D800E sensors but the sensors will out-resolve the lenses about 2:1 at 50 % MTF (estimated). This means they meet the Nyquist frequency sampling criteria.
Thus in almost all cases, the Nikkor will act as the AA filter. ...Show more →
I don't fully agree with Monito on this one. Certainly there will be situations in which a camera without AA filter will produce data that it did not actually see. This includes Moire effects. However, there will also be situations in which the non-AA camera will not produce artifacts but will result in naturally sharper contrast transitions. In these cases there will be no artificial contrast enhancement needed by the AA camera to achieve that acuity at the contrast boundary. There will also be many situations in which neither camera (with and without AA filter) will see or capture sharp contrast situations either because they aren't there in the scene or they aren't in focus.
All of this is largely irrelevant if the user shoots for jpg captures with inappropriate settings that ruin what the camera is capable of capturing in raw files.
I think that the more an image has to be enlarged instead of shrunk to make a required print, the more potential there is for the non-AA camera to produce a better result, so long as artifacts are not present.
Time plus hand-on camera comparisons in a variety of shooting situations will reveal all. Then the speculation and theory will be moot.
p.1 #20 · Sharpen D800 Image or De-moire D800E Image
I did a quick search on the site of the term moire and only a few posts showed up.
I will add that the D800 has loads of moire on its video.
I just took some video at 1080p of a Vizio 42" LCD TV I'm selling locally.
I wanted to show a short video of the TV's functionality to demonstrate to potential customers that it is in fact functional and works 100%.
I took the standard photoshoot style photos of the item using the D800 - that went well, but when it came down the D800 video shoot, that was when things got complicated.
Sound: I wasn't happy with the sound (echo, empty room type audio - just the conditions in the room, probably not the cameras fault). So I attached my Marantz PMD661 XLR solid state recorder to my Rode condenser microphone to 1) test out the attaching an external microphone ability of the d800; and 2) to see how it recorded this inputed audio.
Since the Rode mic I used requires phantom power, I had to use the Marantz PMD661 to enable use of this high quality mic.
The next problem was delivering the audio from the Marantz into the D800's 1/8 stereo mic jack. Fortunately the Marantz has a dedicated RCA L/R out feed so I attached the standard RCA/L-R Stereo to 1/8 stereo cable to deliver the audio from the Marantz to the D800.
Next - adjusting the D800's audio was a bit difficult because of how using the control pad jumps around and how precise you have to be when selecting from Automatic to Manual levels. Note - switching to manual seemed to redirect the D800 to receive audio from its mic input jack. Seems like I was wondering how it knew which audio source to use (internal vs. mic jack). Probably senses a mic plugged in.
So, after all said and done, the audio eventually made its way from the TV to the Marantz to the D800. About at least a 30% boost of quality for the first try.
Video - Moire - Totally present in the video.
The move playback on the LCD TV was then recorded using the D800. Resulting terrible moire across the LCD panel of the TV - yet the parts of the composition not an LCD the frame of the TV, surrounding scene in the D800 video was non-moire effected.
I've basically come to the conclusion that moire is a given under these conditions and to either find another video source, or just mask it in later in post. In any case, it can't be avoided with all practicality on the D800 - as I can see with all the elaborate fixes to the hardware on this.
Video of the moire will eventually be linked through this link I suppose.
